We take cryptocurrency — so the coins you hold can buy real hardware.Pay in crypto, spend it on real hardware. Free insured shipping worldwide over $399. Plain, unbranded boxes, sent from Germany.Free insured shipping over $399. Your rate is locked for 30 minutes once checkout opens.Rate locked 30 minutes. Every item sealed, serial-checked and covered by a 24-month warranty.Sealed, 24-month warranty.

Saved items Sign in
Dual GPU server accelerator card passive cooling 24GB GDDR5 memory

NVIDIA

NVIDIA Tesla K80 24GB GDDR5 dual GPU graphics card

SKU CH-Z455-5E2QS

Dual-GPU server accelerator with 24 GB GDDR5 memory and 480 GB/sec bandwidth for high-throughput compute workloads

  • Memory TypeGDDR5
  • Dual Kepler GK210 GPUs deliver 4992 CUDA cores for massive parallel compute
  • 24 GB GDDR5 memory (12 GB per GPU) supports large datasets in memory
  • 480 GB/s aggregate bandwidth (240 GB/s per GPU) accelerates data-intensive workloads
  • Up to 2.91 TFLOPS double-precision performance with GPU Boost speeds scientific calculations
  • Passive cooling design enables dense server deployments without active airflow per card

Dual-GPU server accelerator for compute workloads

The NVIDIA Tesla K80 pairs two Kepler GK210 GPUs on a single board, delivering 4,992 CUDA cores and 24 GB of GDDR5 memory split evenly at 12 GB per GPU. It targets data-centre tasks that demand high double-precision throughput and large memory capacity.

Passive cooling with up to 2.91 TFLOPS double

The card uses passive cooling and relies on server chassis airflow, so acoustic output is dictated by the host system rather than an on-board fan. Each GPU provides 240 GB/sec memory bandwidth (480 GB/sec aggregate with ECC off) while GPU Boost manages clocks within the thermal envelope.

HP part numbers J0G95A 797637-001 796124-001

These references confirm the exact accelerator variant supplied for compatible HP server platforms. No additional software licences or cables are included; the board is ready for installation in qualified systems that provide sufficient slot power and airflow.

Highlights

  • Dual Kepler GK210 GPUs deliver 4992 CUDA cores for massive parallel compute
  • 24 GB GDDR5 memory (12 GB per GPU) supports large datasets in memory
  • 480 GB/s aggregate bandwidth (240 GB/s per GPU) accelerates data-intensive workloads
  • Up to 2.91 TFLOPS double-precision performance with GPU Boost speeds scientific calculations
  • Passive cooling design enables dense server deployments without active airflow per card

Specifications

Memory TypeGDDR5

Questions about this item

Will this dual-GPU card fit in a standard server chassis?

The card uses passive cooling and is built for server deployment, so it requires a chassis with adequate forced airflow and a full-height, dual-slot PCIe slot.

How much memory does each GPU provide for compute workloads?

Each of the two Kepler GK210 GPUs has 12 GB of GDDR5, giving 24 GB total across the board.

What double-precision performance can I expect from the K80?

The accelerator delivers up to 2.91 Tflops of double-precision compute with GPU Boost enabled.

Also in this aisle

People compared these

Same shelf, same checkout — eight coins and a 30-minute rate lock.